The Model Index
Up to four models

Qwen3.6 27B vs Qwen3.5-27B

On the index (GPT-4 = 100), Qwen3.6 27B scores 141 and Qwen3.5-27B 137. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Qwen3.6 27B
Presets
Qwen3.6 27B
141125–158
WeightsOpen
ReleasedApr 2026
Price in / out$0.32 / $3.2
Context262K
Core tests taken2 of 9
Qwen3.5-27B
137119–156
WeightsOpen
ReleasedFeb 2026
Price in / out$0.30 / $2.4
Context262K
Core tests taken0 of 9

Qwen3.6 27B and Qwen3.5-27B are too close to call (4 points apart, within the uncertainty), at a similar price.

Test by testthird-party results

Qwen3.6 27BQwen3.5-27B
0%25%50%75%100%AGENTSDTBenchDTBenchLMCALMCA
Every result as a table
TestQwen3.6 27BQwen3.5-27B
Agents
DTBench78.1%82.4%
LMCA34.5%34.0%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG